Back to basics: Observed statistics are sufficient to predict drug responses
- 1Tahoe Therapeutics, South San Francisco, CA, USA
Abstract
Predicting how cells, tissues, and patients will respond to a drug, cytokine, or genetic perturbation is central to biological and clinical reasoning. The practical goal is to estimate the analysis-ready readouts that support this reasoning: which cellular responses are context-dependent, which perturbations reveal shared or divergent mechanisms, and which observations should become the basis for the next experiment. Here we introduce Rhaister, a perturbation-response predictor that operates directly on screen-level summary statistics. By measuring just a few perturbations in a new biological context, Rhaister predicts the unmeasured perturbations by learning how response patterns vary across reference contexts. This formulation applies to both fine-grained molecular readouts, such as transcriptional responses from Tahoe-100M or other large perturbation screen, or on phenotypic endpoints. To train and apply Rhaister on pheno-typic endpoints we created Emerald Bay, a purpose-built Tahoe dataset that unifies multi-day cancer drug perturbation, pooled Mosaic tumor contexts, and paired transcriptomic response measurements * . Across these settings, Rhaister matches or exceeds substantially more expensive virtual-cell models, often achieving the highest values possible in evaluation metrics, while training in seconds and running predictions in milliseconds. On Emerald Bay, Rhaister predicts context-specific drug phenotypes from sensitivity measurements alone and improves further when including transcriptomic information. We also introduce Rhaister-O, predicting drug responses in new contexts from baseline expression alone and, to our knowledge, provides the first zero-shot model for this task. Rhaister establishes summary-statistic perturbation modeling as a fast, interpretable framework for predicting biological response across new contexts.
Related
- Electrochemical kinetics, microkinetics modelling -A tutorial from basics to practical implementation
- Structural and Spectroscopic Characterization of Self-Sufficient Cytochrome P450 Fusion Enzymes: CYP102A1 and CYP116B
- Physics-Guided Learning with Calibrated Uncertainty for Molecular Thermodynamics Beyond Observed Chemical Space
- Spectroscopic study of the tunneling dynamics in N2-water observed in the O–D stretch region
- Sliding Charge Density Wave observed through Band Structure
BibTeX
@misc{svensson2026back,
title = {{Back to basics: Observed statistics are sufficient to predict drug responses}},
author = {Valentine Svensson and Umair Khan and Hamed Heydari and Airol A. Ubas and Nicole Thomas and Daniele Merico and Hani Goodarzi and Johnny X. Yu and Nima Alidoust and Shreshth Gandhi},
howpublished = {bioRxiv (Cold Spring Harbor Laboratory)},
year = {2026},
month = jun,
doi = {10.64898/2026.06.09.731197},
url = {https://doi.org/10.64898/2026.06.09.731197},
note = {CC BY},
}
RIS
TY - UNPB
TI - Back to basics: Observed statistics are sufficient to predict drug responses
AU - Valentine Svensson
AU - Umair Khan
AU - Hamed Heydari
AU - Airol A. Ubas
AU - Nicole Thomas
AU - Daniele Merico
AU - Hani Goodarzi
AU - Johnny X. Yu
AU - Nima Alidoust
AU - Shreshth Gandhi
PY - 2026
DA - 2026/06/12
PB - bioRxiv (Cold Spring Harbor Laboratory)
DO - 10.64898/2026.06.09.731197
UR - https://doi.org/10.64898/2026.06.09.731197
AB - Predicting how cells, tissues, and patients will respond to a drug, cytokine, or genetic perturbation is central to biological and clinical reasoning. The practical goal is to estimate the analysis-ready readouts that support this reasoning: which cellular responses are context-dependent, which perturbations reveal shared or divergent mechanisms, and which observations should become the basis for the next experiment. Here we introduce Rhaister, a perturbation-response predictor that operates directly on screen-level summary statistics. By measuring just a few perturbations in a new biological context, Rhaister predicts the unmeasured perturbations by learning how response patterns vary across reference contexts. This formulation applies to both fine-grained molecular readouts, such as transcriptional responses from Tahoe-100M or other large perturbation screen, or on phenotypic endpoints. To train and apply Rhaister on pheno-typic endpoints we created Emerald Bay, a purpose-built Tahoe dataset that unifies multi-day cancer drug perturbation, pooled Mosaic tumor contexts, and paired transcriptomic response measurements * . Across these settings, Rhaister matches or exceeds substantially more expensive virtual-cell models, often achieving the highest values possible in evaluation metrics, while training in seconds and running predictions in milliseconds. On Emerald Bay, Rhaister predicts context-specific drug phenotypes from sensitivity measurements alone and improves further when including transcriptomic information. We also introduce Rhaister-O, predicting drug responses in new contexts from baseline expression alone and, to our knowledge, provides the first zero-shot model for this task. Rhaister establishes summary-statistic perturbation modeling as a fast, interpretable framework for predicting biological response across new contexts.
ER -